Term
| Multiple regression model |
|
Definition
| Using more than one explanatory variable to predict the value of a response variable |
|
|
Term
| Multiple regression model = |
|
Definition
| yi = B0 + B1x1i + B2x2i + … + Bkxki + ei |
|
|
Term
|
Definition
| Shows the linear correlation between each pair of variables under consideration in a multiple regression model |
|
|
Term
|
Definition
| A relationship that exists between two explanatory variables if they have a high linear correlation |
|
|
Term
|
Definition
| The effect of x1 on the value of the response variable does not depend on the value of x2 |
|
|
Term
| Multiple regression, F0 = |
|
Definition
|
|
Term
| Multiple regression, F0 using R^2 = |
|
Definition
R^2/(1-R^2) x [(n-k+1)/k]
where k is the number of explanatory variables |
|
|
Term
| Decision rule for hypothesis testing, multiple regression |
|
Definition
| If P < alpha, then reject the null |
|
|
Term
| 1. Guidelines in developing a multiple regression model |
|
Definition
| Correlation matrix to identify explanatory variables that have a high correlation with the response variable |
|
|
Term
| 2. Guidelines in developing a multiple regression model |
|
Definition
| Use all explanatory variables that have been identified by the researcher |
|
|
Term
| 3. Guidelines in developing a multiple regression model |
|
Definition
| Run regressions, one by one removing variables with small t-statistics and high p-values |
|
|
Term
| 4. Guidelines in developing a multiple regression model |
|
Definition
| Repeat step 3 until all slope coefficients are significantly different from 0 |
|
|
Term
| 5. Guidelines in developing a multiple regression model |
|
Definition
| Draw residual plots to see if the model is appropriate |
|
|
